Scats are dry and inoffensive, and many samples can be obtained from known individuals during a successful live-trapping program. Occasional scats can also be collected in the field, and are useful if they can be correctly identified. By contrast, the stomachs of dead weasels can be collected from trappers, gamekeepers, or game wardens, but they give only one sample per individual and are less pleasant to handle than scats.
Teeth and large pieces of bone discarded in dens can often be identified, but dens are hard to find even when inhabited by weasels wearing radio transmitters. Only in alpine and arctic grasslands can weasel dens be found quite easily, because weasels often commandeer the large and conspicuous overwintering nests made by their vole and lemming prey. By systematicly searching immediately after the spring thaw, finding those nests that have been used during the winter by weasels is relatively easy. Weasels leave behind many easily identifiable rodent teeth and small bones, and often the actual number of voles eaten can be calculated (Fitzgerald 1977; Sittler 1995). In more temperate climates where large prey are available, such as rabbits or seabirds, carcasses may show canine marks that sometimes match the distance between the canines of weasels or stoats (Hewson & Healing 1971; Lyver 2000).
Because metabolism in weasels is so rapid, stomachs contain undigested meat only from very recent meals. Further down the intestine, and in all scats, only hairs, feathers, and bone fragments remain. These can be identified from a combination of minute differences in their structure (Day 1966). The extent to which these differences can be used to identify a weasel's most recent meal depends on the characteristics of the potential prey groups and of the habitat of the study area.
The various species of shrews cannot be separated in scat or gut samples from weasels, even though shrews belong to several different genera. Likewise, rabbits and hares can be grouped only as "lagomorphs," because in most samples they cannot be distinguished reliably. The two most common genera of voles, Cleth-rionomys and Microtus, are represented throughout the northern hemisphere and are among the most frequently eaten prey of all species of weasels. In Britain, remains of bank voles (C. glareolus) in diet samples can be distinguished from those of field voles (M. agrestis) (Day 1966), but in the United States, red-backed voles (C. gapperi) can seldom be distinguished from meadow voles (M. pennsylvanicus) in gut samples (Brown 1952; R.A. Powell, personal observation). Carrion taken from a large carcass, such as predator-killed or road-killed deer or sheep, especially the inner parts without hair, often cannot be detected at all.
The list of potential food items to be identified in a weasel gut is relatively short, since they have evolved as specialist predators of small vertebrates (mammals, birds, and lizards). In their native habitats they eat insects, worms, or vegetable matter (usually only berries) only when extremely hungry. Moreover, since it is fair to assume that weasels never eat hair or feathers except in the course of eating the animal to which they were attached, one can make a rough estimate of the total number of individual prey eaten.
The stomach capacity of most weasels is only about 10 to 20 g, whereas the average weight of a small rodent is about 15 to 30 g. Hence, a weasel cannot eat more than one small rodent at a sitting, so a single stomach, intestine, or scat usually contains the remains of only one item. Conversely, one item can appear in more than one scat, so a group of scats collected at the same time and place has to be treated as a single sample.
The nutritional value of each prey is related to its body size. We can see which items are the most profitable for weasels to hunt and eat by calculating the diet in terms of the weights of the various types of prey eaten rather than their number. The imbalance between large and small items is then corrected, because, for example, seven birds' eggs at 3 g each count the same as two mice at 10 g or one meal of 20 g taken off a dead rabbit. In a sample from which ten birds' eggs, eight mice, and six meals of rabbit were identified, the total weight of prey eaten would be 30 + 18 + 120 = 230 g, of which eggs contributed 13%, mice 35%, and rabbits 52%. That looks quite different from the same data expressed as percentage frequency of occurrence, that is, eggs 42%, mice 33%, and rabbits 25%.
On the other hand, the results of weighting the prey items by body size must be treated with caution, for several reasons. First, the proportion of any one item is, by definition, relative to the total, so the items are not independent. Hence, we cannot say from such figures that weasels from one area rely more on a given type of prey than those from another area. Second, one cannot always discern from their remains whether individual prey were adult or juveniles, which can differ tremendously in size (and in catchability). Third, the results are greatly influenced by how much one assumes a weasel eats from a single large carcass. And fourth, small prey have relatively more bones and hair, which are not equally represented in the remains. Weasels cannot digest hair, so all fur eaten reap pears in the scats (Gamberg & Atkinson 1988), although weasels will avoid eating skin and hair if there is plenty of meat. Large bones will also be avoided, and small ones can be partially digested. Therefore, analyses of hair and bones cannot reflect all prey equally. Even so, calculating the proportions of prey apparently eaten is worth doing to show that, in general, small prey are much less profitable (i.e., return less energy for effort) to eat than large ones.
Identifying a weasel's menu is the easy part of the job. Deciding what the figures mean is far more difficult. To begin with, simple lists of prey eaten by weasels from different places will be biased toward the season, the habitat, and the ages and sex of the weasels most often represented. Likewise, samples of different composition cannot be compared with each other directly, because weasels of different sexes or ages, and in different seasons, may eat different things. For example, some foods, such as birds' eggs, are most available in spring and early summer: It would not be valid to compare the proportion of birds' eggs in samples from two areas unless both had been collected in the same season. And, of course, large samples give much more reliable information than small ones. Any analysis has to reach some compromise between splitting, to avoid compounding different effects in one sample, and lumping, to increase the sizes of the samples.
The large and scattered scientific literature on weasels contains many descriptions of what they eat. Figures 5.1 to 5.4 summarize some of them. Much can be learned about the feeding habits of these little carnivores by setting out the available information systematically in this way, but there are limitations. In the original papers cited here, the data were presented in different ways. Some researchers counted the number of mice that were identified, and expressed the total for mice as a percentage of all the food items identified. Others counted the number of samples containing mice and expressed that as a percentage of all the samples examined. Some included empty stomachs in the figure for total samples, while some excluded them.
Wherever possible, we have standardized the data by recalculating all the results as the number of each item counted as a percentage of all items. This is not, in fact, the best way to compare the diets of weasels in different places, because the proportion of each item is not independent. Pie charts calculated in this way, however, do give a quick and vivid impression of the differences in diets of weasels in different environments, and within broad limits these patterns probably do reflect real variations in how weasels make their living in different kinds of places.
Was this article helpful?